all_results.json

{
    "epoch": 0.9221902017291066,
    "eval_logits/chosen": -0.05567166954278946,
    "eval_logits/rejected": 0.10712358355522156,
    "eval_logps/chosen": -128.39443969726562,
    "eval_logps/ref_chosen": -127.12059020996094,
    "eval_logps/ref_rejected": -163.7115478515625,
    "eval_logps/rejected": -165.0626220703125,
    "eval_loss": 0.6066033840179443,
    "eval_rewards/accuracies": 0.5694444179534912,
    "eval_rewards/chosen": -0.012738817371428013,
    "eval_rewards/margins": 0.0007717205444350839,
    "eval_rewards/rejected": -0.013510540127754211,
    "eval_runtime": 145.1462,
    "eval_samples": 1080,
    "eval_samples_per_second": 7.441,
    "eval_sft_loss": 0.4342755675315857,
    "eval_steps_per_second": 1.86,
    "total_flos": 0.0,
    "train_loss": 0.6159235239028931,
    "train_runtime": 636.1291,
    "train_samples": 1387,
    "train_samples_per_second": 2.18,
    "train_steps_per_second": 0.016
}